[Changes in early auditory evoked potentials in acoustic neuroma].

The auditory brain-stem responses (ABRs) of 38 patients with acoustic neuroma (AN) were recorded pre-operatively and evaluated with regard to pathological deviations from the normal. Several qualitative and quantitative features of the ABRs were compared statistically with the results obtained from a control group for age and sex, and considering patients with various cochlear pathologies (Meniere's disease, sudden deafness, presbycusis). Retrocochlear disorders were ruled out by computed tomography or magnetic resonance imaging. Statistical analysis yielded the following results. The central conduction time (latency difference between Jewett waves J1 and J5) is significantly prolonged in the AN group. A critical value of 4.3 ms separates the retrocochlear from the cochlear lesions, with a sensitivity and specificity of 90%. If conductive hearing loss can be excluded, and if the amount of cochlear hearing loss is taken into account, about the same separation effectively is achieved by evaluation of the sole latency of J5 and its stimulus level dependence. The interaural difference of the central conduction time can be a useful parameter for indicating the presence of acoustic tumours, unless both sides are affected. The same is true of the interaural differences for the J5 latency, but this quantity must be corrected for the effects of peripheral hearing loss. The critical interaural difference values are 0.3 ms for the conduction time (at any stimulus level) and 0.5 ms for the latency of J5 (for stimulus levels above 60 dB nHL). A weaker distinction is provided by the amplitude of J5 and the amplitude ratio of J5 and J1. These quantities are only slightly smaller in cases of AN. Only the combination of small J5 amplitudes with large latency values can be regarded as being characteristic of retrocochlear lesions. This can be highlighted in an amplitude-latency diagram. Further hints as to the site of the lesion can be obtained from a distorted waveform morphology and from an abnormally large distance between hearing threshold and response threshold. The combined evaluation of all these features optimizes the effectivity of ABRs as a powerful tool in the diagnosis of retrocochlear disorders, but some false-positive and false-negative results cannot be excluded entirely.